Discrimination of an amphetamine-pentobarbitone mixture by rats in an AND-OR paradigm.

Rats were trained to discriminate a mixture of amphetamine plus pentobarbitone from either drug separately in a two-bar procedure with food reinforcement. Discrimination was 86% accurate after 48 sessions, and no dose of amphetamine or pentobarbitone alone produced mixture-appropriate responding. Some mixtures increased response rates whereas the same doses of each drug separately had little effect. The same rats were then trained to discriminate a mixture from saline. There was a continuing lack of discriminative response to amphetamine and only a partial response to pentobarbitone, and under these conditions mixtures did not increase overall response rates. Thus, the way rats are trained, and their previous history, can determine the characteristics of the cue obtained.